Thursday, December 23, 2010

Great Moments in Yahoo! Box Scores



It's not just baseball.

Luckily, I saw a couple tweets about this...I don't follow the NBA closely enough to ever click on Rockets/Clippers box scores.

Monday, December 20, 2010

Hitting by Lineup Slot, 2010

I devoted a whole post to leadoff hitters, whether justified or not, so it's only fair to have a post about hitting by batting order position in general. I certainly consider this piece to be more trivia than sabermetrics, but if nothing else it's fun to find new ways to expose the disaster that was Seattle's offense.

The data in this post was taken from Baseball-Reference. The figures for each team's runs are park-adjusted; BA, OBA, and SLG are raw, and OBA is figured as (H + W)/(AB + W). RC is ERP, including SB and CS, as used in my end of season stat posts. The weights used are constant across lineup positions; there was no attempt to apply specific weights to each position, although they are out there and would certainly make this a little bit more interesting.



NL #3 hitters were the most productive in 2009 as well. American League teams had their best hitters in the cleanup spot on average. In the leadoff piece, I touched on how leadoff hitters as a group were below-average; here we see that it was the AL that produced that result, as junior circuit leadoff men were less productive than any other spots other than #8 and #9. #2 hitters continued to be above-average, which is something of a departure from the long-term trend, and certainly is sabermetrically-approved.

Excluding NL #9 hitters (because of the many PA for pitchers), the two least-productive lineup slots were AL #8 and #9. The NL averaged higher RG at slots 1, 3, 5, 6, and 8.

Next, here are the team leaders in RG at each lineup position. The player listed is the one who appeared in the most games in that spot, which is sometimes misleading. For example, Nelson Cruz appeared in 54 games for Texas at #5 to Josh Hamilton's 52. Cruz had a very good 979 OPS in those appearances, but Hamilton turned in a whopping 1178 and was obviously the man most responsible for Texas' superiority at #5.



The only team with two league-leading positions was Oakland; as we'll see below, they also had two league-trailing positions. Keeping in mind that the AL and NL RG averages were 4.45 and 4.33 respectively, every lineup spot had at least one above-average team performance except for NL #9.



Two A's, three Mariners, two Astros, two Dodgers. What really boggles the mind is that Seattle had a spot (5th) in the heart of the order slugging under .300. Their .297 SLG was the better than only six other lineup spots (excluding NL #9 hitters), the highest of which were the PIT/LAA eights...and the SEA sevens. It was also a bad year to put a guy named Lee in the heart of your order as a NL team.

The two charts that follow display the top ten positions based on runs above average. RAA in this case is only in comparison to the AL or NL average at each position. The practical result of this is that NL #3 slots are being compared to a 6.04 RG compared to 5.20 for NL #4s. That doesn't actually mean that if your NL team got 5.5 out of #3 and #4 that the former were hurting and the latter helping. So these figures are presented for fun more than analysis:



As you'll see with the bottom ten, most of the extreme team positions occurred in key lineup slots--3, 4, 5, 1, etc. This makes sense, since those positions are more likely to be manned by one player and there aren't many teams that can go nine deep, so there's not as much variation at the bottom of the order.



There are those Mariner #5s again. The average AL #5 spot hit .264/.329/.437, 5.0 RG to Seattle's .210/.258/.297, 2.5 RG.

Finally, these charts give each team's ranking within their league in RG at each spot. The top and bottom three in each league are highlighted. While the NL has sixteen teams to the AL's fourteen, three still represents the top and bottom 20% in each circuit (rounded to a whole number--for the AL 20% is actually 2.8, for the NL it's 3.2).





The only teams that did not have even a single spot in the top bottom three were San Francisco and Tampa Bay. The Rays were #4 at spots 1-3, so the top of their order was quite productive relative to the league average.

Seattle was the punchline that keeps giving. Their leadoff hitters (almost all Ichiro) were the most productive in the AL, but eleventh was their best showing at any other slot.

Complete data is available in this spreadsheet.

Wednesday, December 15, 2010

Shallow Reflections on Bob Feller

Like many other Cleveland kids of the last couple of generations, I met Bob Feller and got his autograph once. It was at a Discount Drug Mart. Feller made countless such appearances, at drug stores and county fairs and other places where the pay couldn't have possibly been that lucrative. I'm not an autograph/memorabilia aficionado, but I recall reading that Feller's autograph was one of the least valuable for any player of his stature because of the huge supply.

At least in Cleveland, the legend outranked the ballplayer. There's only one player statue outside of Jacobs Field, and it is of course of Bob Feller. Feller is nearly unique in Indians history--a great, Hall of Fame level player that spent his career with just the Tribe. Even counting half-career Indian stars, only Tris Speaker and Nap Lajoie would be able to contest the title of greatest Indian. While Lajoie spent most of his career in Cleveland (65% measured by games played), with the team called the "Naps" for many years, he played before the radio era, before the liveball era, before the team was called the Indians, and never won a pennant in Cleveland.

Thus, Feller's status as a singular franchise icon is arguably unique for one of the sixteen franchises that made up MLB for sixty years. The Yankees (Ruth, Gehrig), Giants (Mathewson, Hubbell), Dodgers (Robinson, Snider), Red Sox (Williams, Yaz), White Sox (Thomas, Fox), Cardinals (Musial, Gibson), Tigers (Cobb, Kaline), Pirates (Wagner, Clemente), Cubs (Banks, Santo), Phillies (Schmidt, Roberts), Reds (Rose, Bench), and Braves (Aaron, Mathews) all have at least two players who, had they been Indians, could compete with Feller for that title (before you complain about omissions from that list, I stopped at two for each team even if there were more, and I'm not suggesting that all of the listed players were better than Feller by any stretch).

The A's might have the situation that most closely parallels the Indians--many of their stars were half-career guys. Still, Foxx, Grove, Simmons, Plank, Cochrane, Collins, Baker...the sheer bulk of icon candidates has to count for something. Among the teams that have used a name change to make a clean break from their past, the Browns/Orioles boast Ripken and Palmer and the Senators/Twins combine for Johnson and Puckett. I might be overselling my case, but I think it's safe to say that if there are any other long-time franchises with a player that is equally Mr. [Nickname] as Feller is Mr. Indian, it's due to that individual's transcendent greatness and not to the dearth of other candidates.

If my thoughts ring a little cold when offered upon the passing of a legend, please reconsider. I said the legend outranked the ballplayer in Cleveland, but I said nothing about the legend outranking the man.

Monday, December 13, 2010

Leadoff Hitters, 2010

This post kicks off a series of posts that I write every year, and therefore struggle to infuse with any sort of new perspective. However, they're a tradition on this blog and hold some general interest, so away we go.

This post looks at the offensive performance of teams' leadoff batters. I will try to make this as clear as possible: the statistics are based on the players that hit in the #1 slot in the batting order, whether they were actually leading off an inning or not. It includes the performance of all players who batted in that spot, including substitutes like pinch-hitters. Listed in parentheses after a team are all players that appeared in twenty or more games in the leadoff slot--while you may see a listing like "MIN (Span) this does not mean that the statistic is only based solely on Span's performance; it is the total of all Minnesota batters in the #1 spot, of which Span was the only one to appear in that spot in twenty or more games. I will list the top and bottom three teams in each category (plus the top/bottom team from each league if they don't make the ML top/bottom three); complete data is available in a spreadsheet linked at the end of the article. There are also no park factors applied anywhere in this article.

That's as clear as I can make it, and I hope it will suffice. I always feel obligated to point out that as a sabermetrician, I think that the importance of the batting order is often overstated, and that the best leadoff hitters would generally be the best cleanup hitters, the best #9 hitters, etc. However, since the leadoff spot gets a lot of attention, and teams pay particular attention to the spot, it is instructive to look at how each team fared there.

The conventional wisdom is that the primary job of the leadoff hitter is to get on base, and most simply, score runs. So let's start by looking at runs scored per 25.5 outs (AB - H + CS):

1. NYA (Jeter/Gardner), 6.5
2. FLA (Coghlan/Maybin/Bonifacio/Ramirez), 6.0
3. DET (Jackson), 5.9
Leadoff average, 5.0
ML average, 4.4
28. CLE (Brantley/Crowe/Cabrera), 4.0
29. WAS (Morgan), 4.0
30. SEA (Suzuki), 4.0

Obviously this category is heavily influence by the quality of the subsequent batters in the lineup; the best indication of this is Ichiro's last-place finish, as you'll see that his leadoff spot actually ranks among the leaders in a couple of more independent categories. Ichiro was the only batter to appear in the leadoff spot in all of his team's games; Juan Pierre (156), Rickie Weeks (155), and Denard Span (151) were the other batters to appear in 150 or games.

The other obvious metric to look at is On Base Average, which speaks to the other conventional goal of a leadoff hitter. The figures here exclude HB and SF to be directly comparable to earlier versions of this article, but those categories are available in the spreadsheet if you'd like to include them:

1. ARI (Johnson/Drew/Young), .366
2. SEA (Suzuki), .358
3. LA (Furcal/Podsednik), .351
Leadoff average, .324
ML average, .322
28. CIN (Phillips/Cabrera/Stubbs), .299
29. WAS (Morgan), .293
30. CLE (Brantley/Crowe/Cabrera), .292

The Reds just cannot seem to find a way to get their leadoff hitters on base. Last year they ranked 29th with a .301 OBA led by Willy Taveras, Drew Stubbs, and Chris Dickerson; and in 2008 they were 24th with Jerry Hairston, Corey Patterson, Jay Bruce and Dickerson. At least this year they weren't wasting PAs on proven failures like Taveras and Patterson.

As alluded to above, Seattle's leadoff hitters had the highest OBA in the American League, yet scored the fewest runs per out.

The next statistic is what I call Runners On Base Average. The genesis of it is from the A factor of Base Runs. It measures the number of times a batter reaches base per PA--excluding homers, since a batter that hits a home run never actually runs the bases.

My 2009 leadoff post was linked to a Cardinals message board, and this metric was the cause of a lot of confusion (this was mostly because the poster in question was thick-headed as could be, but it's still worth addressing). ROBA, like several other methods that follow, is not really a quality metric, it is a descriptive metric. A high ROBA is a good thing, but it's not necessarily better than a slightly lower ROBA plus a higher home run rate (which would produce a higher OBA and more runs). Listing ROBA is not in any way, shape or form a statement that hitting home runs is bad for a leadoff hitter. It is simply a recognition of the fact that a batter that hits a home run is not a baserunner.

ROBA also removes CS, and so the formula is (H + W - HR - CS)/(AB + W):

1. SEA (Suzuki), .337
2. NYA (Jeter/Gardner), .328
3. DET (Jackson), .323
5. LA (Furcal/Podesdnik), .323
Leadoff average, .296
ML average, .290
26. TOR (Lewis/Snider/Wise), .276
28. SF (Torres/Rowand), .269
29. CIN (Phillips/Cabrera/Stubbs), .267
30. WAS (Morgan), .260

Arizona's leadoff hitters, who led the majors in OBA, rank sixth in ROBA because they were second in the majors with 25 homers (Milwaukee hit 28 from the leadoff spot). Washington loses ground on the list (although they were just 29th in OBA) not because their leadoff hitters were driving the ball out of the park (5 homers ranked tied for fifth-fewest), but because they led the majors by being caught stealing 19 times.

I will also include what I've called Literal OBA here--this is just ROBA with HR subtracted from the denominator so that a homer does not lower LOBA, it simply has no effect. You don't really need ROBA and LOBA (or either, for that matter), but this might save some poor message board out there twenty posts, so here goes. LOBA = (H + W - HR - CS)/(AB + W - HR):

1. SEA (Suzuki), .340
2. NYA (Jeter/Gardner), .332
3. DET (Jackson), .330
4. ARI (Johnson/Drew/Young), .330
Leadoff average, .301
ML average, .298
26. CLE (Brantley/Crowe/Cabrera), .280
28. SD (Hairston/Venable/Gwynn/Eckstein), .275
29. CIN (Phillips/Cabrera/Stubbs), .273
30. WAS (Morgan), .262

The next two categories are most definitely categories of shape, not value. The first is the ratio of runs scored to RBI. Leadoff hitters as a group score many more runs than they drive in, partly due to their skills and partly due to lineup dynamics. Those with low ratios don’t fit the traditional leadoff profile as closely as those with high ratios:

1. FLA (Coghlan/Maybin/Bonifacio/Ramirez), 2.5
2. TEX (Andrus), 2.4
3. DET (Jackson), 2.1
Leadoff average, 1.7
27. KC (Podsednik/Blanco/DeJesus), 1.4
28. SD (Hairston/Venable/Gwynn/Eckstein), 1.4
29. SF (Torres/Rowand), 1.3
30. PHI (Victorino/Rollins), 1.2
ML average, 1.1

Florida's leadoff hitters scored a lot of runs (as we saw earlier), so it's no surprise they had a high R/RBI ratio. Texas ranks second because they drove in just 42 runs (tied with Cleveland for fewest), and with a ML-low .290 SLG it's not hard to see why (they trailed in SLG by a wide margin; CHA was next at .310).

A similar gauge, but one that doesn't rely on the teammate-dependent R and RBI totals, is Bill James' Run Element Ratio. RER was described by James as the ratio between those things that were especially helpful at the beginning of an inning (walks and stolen bases) to those that were especially helpful at the end of an inning (extra bases). It is a ratio of "setup" events to "cleanup" events. Singles aren't included because they often function in both roles.

Of course, there are RBI walks and doubles can be a great way to start an inning, but RER classifies events based on when they have the highest relative value, at least from a simple analysis:

1. CHA (Pierre), 4.1
2. TEX (Andrus), 2.4
3. HOU (Bourn/Bourgeois), 2.6
Leadoff average, 1.1
23. TOR (Lewis/Snider/Wise), .9
ML average, .8
28. NYN (Reyes/Pagan), .7
29. ATL (Prado/Infante), .7
30. SF (Torres/Rowand), .6

The influence of stolen bases is pretty strong in RER, which is why the White Sox rank so highly--their 67 swipes led all leadoff spots, with the Astros next at 61. Atlanta's 10 steals only beat out two teams (Boston and St. Louis) that stole nine.

Speaking of stolen bases, I decided it would be worthwhile this year to look at a pure measure of base stealing. Obviously there's a lot more that goes into being a leadoff hitter than simply stealing bases, but it is one of the areas that is often cited as important. So I've included the ranking for what some analysts call net steals, SB - 2*CS. I'm not going to worry about the precise breakeven rate, which is probably closer to 75% than 67%, but is also variable based on situation. The ML and leadoff averages in this case are per team lineup slot:

1. OAK (Crisp/Davis/Pennington), 45
2. CHA (Pierre), 33
3. HOU (Bourn/Bourgeois), 31
Leadoff average, 11
ML average, 3
28. ATL (Prado/Infante), -2
29. LAA (Aybar/Abreu), -4
30. COL (Fowler/Gonzalez/Young), -10

It is really quite mind-boggling that a playoff contender playing in the major's most offense-friendly park would allow its leadoff men to attempt 41 steals with a 59% success rate. And there's those Moneyball A's...never mind, too easy.

Oakland and Philadelphia tied for the lead in SBA at 87.5%; the Phillies were 35/40, and fourth in net steals. The only teams below the 2/3 success rate for net steals were BOS (9/14), LAA (22/35), ATL (10/16), and of course COL (24/41). Leadoff hitters composite SBA was 75.7%, compared to the overall major league rate of 72.4%.

Let's shift gears back to quality measures, beginning with one that David Smyth proposed when I first wrote this annual leadoff review. Since the optimal weight for OBA in a x*OBA + SLG metric is generally something like 1.7, David suggested figuring 2*OBA + SLG for leadoff hitters, as a way to give a little extra boost to OBA while not distorting things too much, or even suffering an accuracy decline from standard OPS. Since this is a unitless measure anyway, I multiply it by .7 to approximate the standard OPS scale and call it 2OPS:

1. ARI (Johnson/Drew/Young), 850
2. LA (Furcal/Podsednik), 794
3. MIL (Weeks), 791
4. SEA (Suzuki), 776
ML average, 733
Leadoff average, 722
28. SD (Hairston/Venable/Gwynn/Eckstein), 638
29. WAS (Morgan), 633
30. CLE (Brantley/Crowe/Cabrera), 629

Along the same lines, one can also evaluate leadoff hitters in the same way I'd go about evaluating any hitter, and just use Runs Created per Game with standard weights (this will include SB and CS, which are ignored by 2OPS):

1. ARI (Johnson/Drew/Young), 6.2
2. LA (Furcal/Podsednik), 5.4
3. MIL (Weeks), 5.3
4. SEA (Suzuki), 5.2
ML average, 4.5
Leadoff average, 4.4
28. TEX (Andrus), 3.1
29. CLE (Brantley/Crowe/Cabrera), 3.1
30. WAS (Morgan), 3.1

Not surprisingly, this list is extremely similar to the 2OPS list.

Finally, allow me to close with a crude theoretical measure of linear weights supposing that the player always led off an inning (that is, batted in the bases empty, no outs state). There are weights out there (see The Book) for the leadoff slot in its average situation, but this variation is much easier to calculate (although also based on a silly and impossible premise).

The weights I used were based on the 2010 run expectancy table from Baseball Prospectus. Ideally I would have used multiple seasons but this is a seat-of-the-pants metric. Here are the relevant lines of that RE table:



To calculate the value of a single or walk in (---, 0), simply subtract .492 from .859 to get .367. Similarly, the value of a double is 1.101 - .492 = .610 and a triple is 1.358 - .492 = .866. A home run is worth one run, as the state remains (---, 0) but there is a run on the board.

Assuming (conveniently and inaccurately) that all stolen base attempts occur with 0 out and are of second base, the value of a steal is 1.101 - .859 = .242, which necessarily is the same as the extra value of a double over that of a single or walk.

I will deal with outs in such a manner so as to force the average leadoff hitter to zero RAA. They will not come out to zero without special treatment; after all, this is a theoretical construct. Leadoff hitters are not perfectly average nor are events evenly distributed across base/out states.

First, a caught stealing costs the team the value of the baserunner previously earned (-.367), plus the cost of the out itself, which also applies to (AB - H). So to calculate the out value, solve this equation for x:

0 = .367(S + W - CS) + .61D + .866T + HR + .242SB + x*(AB - H + CS)

For 2010 leadoff hitters, x = -.230, and so our theoretical leadoff RAA (which I'll call raw Leadoff Efficiency because I was already using that name for a different metric in the past) is:

rLE = .367(S + W) + .61D + .866T + HR + .242SB - .583CS - .23(AB - H)

To convert this to a rate (it is a RAA total in its current form), I divided by PA (AB + W) and multiplied by the average number of PA for leadoff hitters in 2010 (742). This yields Leadoff Efficiency:

1. ARI (Johnson/Drew/Young), 27
2. LA (Furcal/Podsednik), 15
3. MIL (Weeks), 15
4. SEA (Suzuki), 12
ML average, 1
Leadoff average, 0
28. SD (Hairston/Venable/Gwynn/Eckstein), -19
29. CLE (Brantley/Crowe/Cabrera), -21
30. WAS (Morgan), -22

The fact that this list is very similar to the lists based on metrics designed to apply generic weights to all batters illustrates how the relative values of offensive events are fairly stable.

One thing I noticed when writing this article was how many teams were using multiple players in their leadoff spot. Compared to 2007-2009, there were indeed a lot of different players used in the role:



The first column is the average number of games in the leadoff spot for the team leader; the second column is the number of teams that had a player appear in 100 or more games a leadoff man, and the third is the total number of players with 20 or more appearances in the leadoff spot.

What was unusual in 2010 was not the number of players appearing in twenty or more games, but rather the lack of players that lead off in the bulk of their team's games. For now it is just a blip; it will be interesting to see if it remains that, or is indicative of a trend. My guess is the former, but it caught my eye and so I mentioned it here.

Assuming for the sake of discussion that it is the beginning of a trend, one would have to question whether the new approach is working. Leadoff hitters' composite OBA was just two points better than the major league average, the smallest margin since I started tracking it in 2005 (the previous low was six points in 2006). Leadoff hitters were also below-average in a generic RC analysis (4.4 RG versus a ML average of 4.5), and it's tough to believe that represents optimal lineup construction.

Here is a link to a Google Spreadsheet with the data used in this post.

Monday, December 06, 2010

Statistical Meanderings, 2010

This is about as close as I get to writing a Jayson Stark-style piece throughout the course of the year. Sine I hate that format, hopefully there will be something of greater interest than sheer trivia here. Most of the statistics mentioned come from my End of Season Stats and are explained in that post:

* Last year the AL/NL scoring gap in terms of R/G was the largest it had been since 1998; this year, at .12 (4.45 to 4.33) it was the narrowest it had been since 1990 (4.30 to 4.20). The overall scoring average of 4.38 was the lowest for the majors since 1992 (4.12); 1992 was also the last time that either of the leagues individually had as low of a scoring rate.

The offensive difference between the AL and NL was largely due to a difference in league batting averages. As a group, the AL and NL had nearly identical walk rates (.095 and .096 walks/at bat) and isolated power (.147 to .144), but the AL BA was five points higher (.260 to .255). The NL slugged just .399, the first sub-.400 league figure since the 1993 NL.

* I list two different winning percentage estimators in my team report. EW% is based on actual runs scored and allowed, while PW% is based on runs created and runs created allowed. Teams whose actual W% were very similar to both of the estimates included (W%, EW%, PW%): Atlanta (.562, .567, .564), Cincinnati (.562, .567, .564), Florida (.494, .501, .495), and Texas (.556, .564, .557).

An interesting group of teams is those whose PW% tracked their actual W% much better than EW% did. These are teams that may be over/underrated for 2011 by those that put a great deal of stock in Pythagorean record as an indicator. Such people are largely strawmen, but regardless, some of the teams in this group are Baltimore (.407, .386, .411), the Cubs (.463, .447, .468), Pittsburgh (.352, .324, .351), and St. Louis (.531, .564, .542). I'll leave it to the reader to find the more conventional Pythagorean watch teams, those whose EW% and PW% are in general agreement and diverge from actual W%.

* Last year, SF games were the lowest scoring in MLB at 7.83 RPG, which was the lowest figure since the 2003 Dodgers. In 2010, 7.83 RPG would have ranked just third-lowest, as Seattle (7.48) and San Diego (7.69) each exhibited a lower scoring context. The Mariners' 7.48 still couldn't touch the 2003 Dodgers at 6.98, but it was the lowest RPG for an AL team since the 1981 Yankees (7.14). Of course, Seattle's RPG was lowered by the .97 park factor, but even after park-adjusting the figure to 7.71, it was still the lowest AL scoring context since the 1989 Angels (7.70, not park-adjusted).

Commenting on Seattle's offensive ineptitude can be considered hitting after the whistle at this point, but allow me to indulge. Their 3.17 runs/game was the lowest since the 1981 Blue Jays averaged 3.10. Seattle's 2.95 R/G at home was the lowest since 1972, when both the Padres (2.71) and Angels (2.76) scored less. While Safeco had a large home/road split in 2010, the five-year PF is .97--a pitcher's park, yes, but not an extreme one.

They were more respectable on the road, averaging 3.38 R/G, a mark which the Pirates (3.14) managed to keep from being even the lowest in 2010 (although outside of the Pirates' showing, it was the fewest since the 1994 Pirates scored 3.20 away from home. I did not run the numbers relative to league average, but it probably wouldn't do to much to help Seattle; while the AL's average of 4.45 R/G is low relative to recent seasons, it's still a perfectly normal league scoring level in historical context.

* Unfortunately, we never got to see the playoff matchup between New York and Tampa Bay. While the concerns about the Rays running wild in such a series were likely overstated, it is true that the Yankees struggled at controlling the stolen base game. The 85.2 SBA against them was easily the highest in MLB, with the Red Sox (80.1) next. The Cardinals lead baseball with only 58.9% of opposition attempts successful.

* It will come as no surprise that Pittsburgh had a terrible defense in 2010. The degree of their anti-dominance may be a little jarring though: last in BA (by ten points, .283), last in OBA (by eight points, .347), last in SLG (by fourteen points, .451). The only team offense that exceeded any of the Pirates' allowed figures was Toronto, which slugged .456.

The Pirates were also last in innings/start (by .11 innings, 5.38), starters' eRA (by a whopping .62 runs, 5.86), and DER (.659). Their bullpen ranked only fifth-worst in eRA (4.82), and their modified fielding average was third-worst (.962). All of this predictably resulted in allowing 5.4 R/G (more than any team managed to score).

* In 2009, playoff teams averaged +72 runs above average on offense and just +44 on defense. In 2010, the teams exhibited more balance, as you can see:



I'd usually snark about defense winning championships at this point.

* You're probably aware that the long-term trend in MLB, pretty much dating all the way back to 1871, has been for fielding averages to increase. For the most part this holds, but there was an odd blip in 2009. The all-time high ML mFA is .9704, set in 2007. In 2008, the mFA rounded to four decimal places was the same but actually was a bit lower. In 2009, however, mFA dropped all the way to .9669, the lowest since 2001. The decline of .36% was the largest in the post-war era.

In 2010, mFA rebounded to .9693, an increase of .25%. That is still the lowest average (excluding 2009) since 2004. I am not claiming that fielding average is an important metric, or that there is a meaningful explanation for the fluctuations, but in looking at league fielding totals it caught my eye.



* Major league teams had a .559 W% at home in 2010, the highest mark since 1978 (.573). 93% of major league teams (28/30) had better records at home than on the road, which sounds like a lot, but while high it isn't extraordinary. (San Diego had the same record home and away). The average for 1961-2010 is 83%, but as recently as 2007-08, 29/30 teams have had better home records. In both 1978 and 1989 all teams had better records at home.

Much was made about the Pirates' .210 road W% (17-64), the worst since the identical showing by the 1963 Mets. Also notable was Detroit's home/road split of .642/.358, which was of equal magnitude to that of the Pirates and was the largest by a .500 or better team since the 1996 Rockies (.679/.346). The Rockies and Braves chipped in to make it four of the 23 highest differentials since 1961 in 2010.

* Cleveland fans seem to be pretty happy with new closer Chris Perez, and given his performance (7th in the AL among relievers with 20 RAR), but it would be a mistake to assume that he's proven himself as the long-term answer at the end of the game. He allowed a low .234 %H, so his 3.72 dRA is well above his 2.10 RRA or 2.86 eRA. The batted ball metrics are even less impressed--4.45 cRA, 4.74 sRA.

* Jonathon Papelbon pitched 67 innings with a 4.24 RRA, which results in 5 RAR; Scott Atchison pitched 60 innings with a 4.16 RRA, for 5 RAR as well. Of course, the similarities end there, as Papelbon's peripherals were much better than Atchison's, but it's never a good thing when your pricy closer is no more effective than the seventh man out of the pen.

* Bobby Jenks has been non-tendered, and obviously I have no insight to offer on his health or his PitchF/x data or anything like that. What I can tell you is that his peripherals were pretty good in 2010: 2.89 dRA (his %H was very high at .365), 3.34 cRA, 2.91 sRA. If he's healthy, he might be a good buy.

* Chad Qualls gave up a massive .397 %H; he actually looks serviceable in dRA (4.26) and the batted ball metrics (4.33 and 3.89).

* Trevor Hoffman was terrible in his swan song, but at least he was consistent across the board in RA estimators: 5.95 RRA, 5.82 eRA, 5.89 dRA, 5.45 cRA, 6.10 sRA. Ryan Madson was consistent in a good way: 2.47, 2.80, 2.88, 2.83, 2.90.

* Last year I pointed out that Francisco Rodriguez didn't pitch very well in the first year of his big contract, so I feel obligated to point out that he was pretty good in his 57 innings in 2010: 2.34 RRA, 3.16 eRA, 3.10 dRA, tied for fifteenth among NL relievers with 16 RAR. Of course, his off-the-field performance took a corresponding nose dive...

* Who is Wilton Lopez? I probably saw less Houston games than any other team this season, so I never saw him pitch. The 27 year-old Nicaraguan rookie ranked among the ten most valuable relievers in the NL (not considering leverage) with a 2.02 RRA in 67 innings and solid, consistent peripherals (3.41 eRA, 3.24 dRA, 3.24 cRA, 3.22 sRA). He was shelled last season in 19 innings, and his strikeout rate (6.7) leaves a lot to be desired, and for the season as a whole he wasn't trusted by Brad Mills, with a below-average Leverage Index. He did inherit .49 runners/appearance, but sometimes a high IR/G goes hand-in-hand with a mop-up role. As if that wasn't enough cold water, his minor league numbers don't look like anything special from a quick glance. It was a nice season in any event.

* I don't list Inherited Runs and Bequeathed Runs Saved on the reports themselves, but if you download the spreadsheets, they are included. The AL leaders in IRSV were Matt Thornton (5.7), Randy Choate (5.1), and Joaquin Benoit (5.0). Dan Wheeler (4.0) also had a particularly good showing from Tampa's pen. Eddie Bonine trailed the AL at -8.4

Among AL Relievers, Dusty Hughes got the least support from subsequent relievers as 15/25 scored (7.2 BRSV). Lance Cormier benefited the most as only 1/30 bequeathed runners came around to score (-8.1 BRSV).

In the NL, Wilton Lopez (9.2), Javier Lopez (9.2), and Santiago Casilla (8.7) were the leaders in stranding runners; Lopez allowed just 1/33 to score. A pair of Dodgers were the trailers: George Sherrill (-5.7) and Ramon Troncoso (-10.5 on 22/37).

Apparently Ronald Belisario was one of their victims, as he got the least support from subsequent relievers (4.6 BRSV on 12/24). Joe Thatcher was the most fortunate, as his Padre penmates prevented all 35 runners he bequeathed from scoring (-10.5).

* The flip side to the last bullet point is bequeathed runs saved for starting pitchers. In the AL, Rich Harden got the most support (2/20, -4.2) followed by Jake Westbrook and Jon Lester (both -3.7). Jason Vargas got the least help (12/20, 6.0).

In the NL, Jonathan Sanchez (3/25, -4.9) and a pair of Braves (Derek Lowe, -4.3 and Tommy Hanson, -3.0) were the best-supported by their pens. Scott Olsen (10/16, 5.0), Chris Narveson (4.8), and Kevin Correia (4.6 despite pitching for San Diego with their excellent bullpen) were the least supported.

* Presented without comment: Max Scherzer 37 RAR, Edwin Jackson 25. Clayton Richard 31 RAR, Jake Peavy 17.

* I usually only include pitchers with 15 starts on the starters report, but I had to throw Stephen Strasburg into the mix. Among NL pitchers with 15 starts (plus himself), he ranked twelfth in RA, fourth in eRA, and first in dRA, cRA, and sRA. Obviously that was in just 68 innings, but it was fun while it lasted.

* It's a shame there is no LVP award, as it would be a runaway in both leagues--Ryan Rowland-Smith, -26 RAR in the AL and Charlie Morton -31 in the NL. Rowland-Smith was 1-10 with a 7.83 RRA in 109 innings, and none of his peripheral RAs were much better (6.37 sRA was his best). He was last in the AL in all of the run averages, plus QS% (20).

Morton's season was more respectable, as he pitched 80 innings and gave up a .367 %H, which means his dRA (5.50) and batted ball RAs (5.10 and 4.56) weren't terrible. Teammate Zack Duke was next on the RAR trailer list (-20) and matched Morton with -42 RAA thanks to being allowed to pitch 159 innings.

* Cliff Lee and Jon Lester are very close in most of the categories I list, in addition to both having names that start with "L" and being left-handed. Lee pitched 4 1/3 more innings, with essentially the same RRA (3.55 to 3.52), eRA (3.19), and cRA (3.55 to 3.52). Lee was a little better in dRA (3.08 to 3.26), Lester better in sRA (3.77 to 3.36). Lee's %H was .300 to Lester's .295; Lee made 106 pitches/start, Lester 105; Lee made 64% quality starts, Lester 63%. Both get credit for 21 RAA, while Lee gets one more RAR (51 to 50).

* The bottom 12 starting pitchers in the AL in RAR are: Ryan Rowland-Smith, Brian Bannister, David Huff, Scott Kazmir, Rich Harden, Josh Beckett, Scott Feldman, Jamie Shields, Nick Blackburn, Jeremy Bonderman, Tim Wakefield, and AJ Burnett. The NL's bottom dozen is a much more conventional list of lousy pitchers: Charlie Morton, Zach Duke, Kyle Lohse, Nate Robertson, Manny Parra, Jeff Suppan, Paul Maholm, Bud Norris, Kevin Correia, Craig Stammen, Dave Bush, and John Ely.

* We all know that Texas owes its success largely to pitchers going deep in games, right? Wait, they only averaged 5.87 IP/S, ninth-lowest in the majors? Well, surely that must be because of some bad starts from Rich Harden or something.

Well, here are the P/S (counting relief appearances as half-starts) for the Rangers' starters with 15 or more starts (including starts made for other teams in the case of Cliff Lee): Lee 106, Wilson 104, Lewis 103, Feldman 95, Harden 93, Hunter 85.

Tampa Bay, their playoff opponents: Price 107, Garza 101, Shields 100, Davis 96, Niemann 89.

A few other teams: White Sox: Danks 106, Peavy 100, Buehrle 100, Floyd 97, Garcia 88.

Boston: Lackey 109, Lester 105, Matsuzaka 105, Beckett 103, Buchholz 100, Wakefield 85

Oakland: Gonzalez 102, Cahill 100, Mazzaro 99, Sheets 98, Braden 95, Anderson 95

Angels: Weaver 109, Santana 108, Saunders 100, Pineiro 100, Kazmir 98

Detroit: Verlander 113, Scherzer 106, Porcello 96, Galarraga 95, Bonderman 93

Of course, facts aren't particularly important when you need to rhetorically twist a playoff series into a morality play.

* Earlier in the season, I posted about the number of no-hitters thrown in 2010 and how it compared to expectation based on both the historical frequency of no-hitters and a theoretical probability of a no-hitter. The details are in that post and will not be repeated here.

There were six no-hitters pitched in 2010 out of 4,924 possible games (double-counting games since of course each pitcher has an opportunity to throw a no-hitter). Given the long-term frequency of no-hitters figured by Tom Flesher (.06%), we'd have expected 2.954 no-hitters. The Poisson distribution yields this expected distribution:



I also offered a distribution based on the theoretical probability derived from the overall major league BA (.2573) with 2.734 expected no-hitters (.056%):



Whichever model you choose, there's nothing particularly shocking about six no-hitters in one season, something that has about a 6% chance of occurring by chance. If I wanted to get cute, I'd point out that 6% is about once every twenty years, and the last big no-hitter season was in 1990, but that would be sabermetric malpractice.

* Speed Score leaders and trailers by position (with the caveat that these are based on just 2010 data when Bill James intended them to be based on multiple years):

C: Miguel Olivo (6.1)/Chris Snyder (.7)
1B: Kevin Youkilis (5.5)/Adrian Gonzalez (1.0)
2B: Sean Rodriguez (6.2)/Luis Valbuena (1.7)
3B: Evan Longoria (5.4)/Wilson Betemit (.9)
SS: Rafael Furcal (7.8)/Juan Uribe (2.2)
LF: Carl Crawford (9.1)/Pat Burrell (1.0)
CF: Dexter Fowler (9.1)/Torii Hunter (2.5)
RF: Will Venable (8.6)/Ryan Ludwick (1.9)
DH: Johnny Damon (5.5)/Willy Aybar (.7)

* RGs for Yankee hitters with 300+ PA: 6.8, 6.1, 5.9, 5.8, 5.6, 5.4, 5.2, 4.3, 4.1. Without doing any formal checks, I have to assume that's one of the more balanced league-leading lineups you are likely to see. The Yankees led in team RG with 5.14; the Red Sox were second at 5.08. Their breakdown for 300+ PA hitters was: 7.6, 6.7, 6.5, 6.2, 5.9, 5.1, 4.9, 4.9, 4.2. The range is only about one run wider, but it's more top-heavy.

* Ryan Braun and Prince Fielder had different shapes to their production (Braun had a .305 BA and .289 SEC while Fielder was at .263/.409), but the outcomes were very similar. Braun created 110 runs while making 434 outs for 6.45 RG, while Fielder created 109 runs in 427 outs for 6.49 RG; both were at +36 HRAA. Position adjustments put Braun several runs ahead in the categories where they are included, but it would be tough to find two stars on a team better matched.

* Park-adjusted stats: Matt LaPorta .222/.307/.364, 3.84 RG, 0 RAR. Justin Smoak .215/.305/.365, 3.87, 0. These two make an interesting pairing since both were the centerpiece of a trade of a former Indians left-handed ace. Of course it has been two years since LaPorta was traded to Cleveland while Smoak was traded this summer, but they both will need to improve on those performances to avoid the trades becoming Santana II and III. Personally, I think Smoak is a much better bet--he's younger and LaPorta was nagged by several injuries during 2010.

* The Indians got nearly identical offensive production out of their two primary center fielders, Trevor Crowe and Michael Brantley. Crowe hit .252, Brantley .247. Each had a .299 OBA. Crowe slugged .334, Brantley .328. Add in steals and they both created 3.42 runs/game and were essentially replacement level (1 RAR each). This was actually an improvement over Grady Sizemore, who hit even worse (.212/.264/.291) in his 137 plate appearances.

* Alex Rodriguez had the worst full-time season of his career in 2010. His hitting relative to the league average was a little worse in 1997, but he was playing shortstop at the time and had an additional 40+ plate appearances than he did in 2010. This is not meant as a condemnation of A-Rod in any way, as he was still a valuable asset and after all is 35 years old.

However, I was surprised by the lack of media excitement about this. Certainly it was pointed out that he was not his old self, but usually anything negative about the man is blown completely out of proportion, and the media could have had a field day if they'd framed the story in a certain light. Why didn't they? Speculation about motivation aside, RBI offer a statistical explanation. Rodriguez batted in 125 runs, most since his MVP campaign in 2007, and the sixth-highest total of his career.

At the same time, Rodriguez' ratio of RBI to RC (using no park adjustment in the latter) was the highest it had ever been at 125/92 = 1.38. His previous full-season high had come in 1999 (111/102, 1.09). For his career, A-Rod's RBI and RC are very close (1855 RBI, 1831 RC), and so he is not a hitter that has a RBI > RC tendency.

Among batters with 300 or more PA, only Pedro Feliz (40/26, 1.51) and Willy Aybar (43/30, 1.45) had a higher ratio than Rodriguez. Ichiro turned in the major's lowest ratio (43/91, .47). A-Rod's relative lack of production was masked by his RBI count, and if it has anything to do with preventing a media freakout, I'm personally pleased it was.

Matt Klaassen of Fangraphs has written about ARod's RC/RBI ratio as well.

* I hadn't noticed how poorly AL shortstops hit until the Silver Sluggers were announced. When I heard "Alexei Ramirez", I was little surprised. Of course, then I looked at the numbers a little more closely, and it's ugly. These are all AL players with 300+ PA who were primarily shortstops:



None of them managed to match even the AL average RG, which leads to this amusing chart of Silver Slugger winners together with their HRAA:



Admittedly, using average as a baseline is a cheat for shock value in this case.

* Another hitter of historical note whose 2010 wasn't up to his own previous standards was Albert Pujols, at least if you believe some accounts. It has been seemingly common to claim that 2010 was Pujols' worst season, but I beg to differ.

Looking at unadjusted (for league or park) RG, it was only Pujols' fourth-worst season, as he had lower RC rates in 2001, 2002, and 2007. Factoring in league and park, I have his ARG at 206, which was actually a smidge better than his pre-2010 career mark of 204, and puts the season well ahead of the three years already discussed and in the same general pool as 2004-2006. Factor in that Pujols set a career high with 690 PA, and my fielding-less WAR pegs it as the fifth-best season of his career--right in the middle. And still a season strong enough to be worthy of NL MVP honors.